| |
Tony's Home (Site not responding. Last check: 2007-10-23) |
 | | Whenever I used the term "parallelizable" for a manifold, I should have said |
 | | "parallelizable with a pseudo-Riemannian metric, invariant under the flat connection naturally associated with the parallelization, whose geodesics are the same as those of that connection". |
 | | For instance, Kervaire [2] proved that a product of spheres is parallelizable as long as at least one of them has odd dimension; most such products are not diffeomorphic to products of Lie groups, since a compact, simply connected Lie group has nontrivial third cohomology. |
| www.valdostamuseum.com /hamsmith/tsemailaddress.html (282 words) |